A Double Extrapolation Primal-Dual Algorithm for Saddle Point Problems

被引:8
|
作者
Wang, Kai [1 ]
He, Hongjin [2 ]
机构
[1] Nanjing Univ Sci & Technol, Sch Sci, Dept Math, Xiaolingwei 200, Nanjing 210094, Peoples R China
[2] Hangzhou Dianzi Univ, Dept Math, Sch Sci, Hangzhou 310018, Peoples R China
基金
中国国家自然科学基金;
关键词
Saddle point problem; Primal-dual algorithm; Extrapolation; Linear convergence rate; Image deblurring; Image inpainting; ALTERNATING DIRECTION METHOD; LINEAR CONVERGENCE; VARIATIONAL-INEQUALITIES; CONVEX-OPTIMIZATION; MINIMIZATION; MULTIPLIERS; FRAMEWORK;
D O I
10.1007/s10915-020-01330-w
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
The first-order primal-dual algorithms have received much considerable attention in the literature due to their quite promising performance in solving large-scale image processing models. In this paper, we consider a general saddle point problem and propose a double extrapolation primal-dual algorithm, which employs the efficient extrapolation strategy for both primal and dual variables. It is remarkable that the proposed algorithm enjoys a unified framework including several existing efficient solvers as special cases. Another exciting property is that, under quite flexible requirements on the involved extrapolation parameters, our algorithm is globally convergent to a saddle point of the problem under consideration. Moreover, the worst case O(1/t) convergence rate in both ergodic and nonergodic senses, and the linear convergence rate can be established for more general cases, where t counts the iteration. Some computational results on solving image deblurring, image inpainting and the nearest correlation matrix problems further show that the proposed algorithm is efficient, and performs better than some existing first-order solvers in terms of taking less iterations and computing time in some cases.
引用
收藏
页数:30
相关论文
共 50 条
  • [31] General Inexact Primal-Dual Hybrid Gradient Methods for Saddle-Point Problems and Convergence Analysis
    Wu, Zhongming
    Li, Min
    ASIA-PACIFIC JOURNAL OF OPERATIONAL RESEARCH, 2022, 39 (05)
  • [32] Primal-dual Interior Point Algorithm for Linear Programming
    Yong, Longquan
    PROCEEDINGS OF FIRST INTERNATIONAL CONFERENCE OF MODELLING AND SIMULATION, VOL II: MATHEMATICAL MODELLING, 2008, : 432 - 436
  • [33] A hybrid primal-dual algorithm with application to the dual transportation problems
    Choi, G
    Kim, C
    COMPUTATIONAL SCIENCE AND ITS APPLICATIONS - ICCSA 2005, VOL 4, PROCEEDINGS, 2005, 3483 : 261 - 268
  • [34] A primal-dual interior-point algorithm for nonlinear least squares constrained problems
    M. Fernanda
    P. Costa
    Edite M. G. P. Fernandes
    Top, 2005, 13 (1) : 145 - 166
  • [35] A primal-dual method with linear mapping for a saddle point problem in image deblurring
    Xie, Zhipeng
    JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2017, 42 : 112 - 120
  • [36] Random extrapolation for primal-dual coordinate descent
    Alacaoglu, Ahmet
    Fercoq, Olivier
    Cevher, Volkan
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 119, 2020, 119
  • [37] A Smooth Double Proximal Primal-Dual Algorithm for a Class of Distributed Nonsmooth Optimization Problems
    Wei, Yue
    Fang, Hao
    Zeng, Xianlin
    Chen, Jie
    Pardalos, Panos
    IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2020, 65 (04) : 1800 - 1806
  • [38] An Inexact Primal-Dual Smoothing Framework for Large-Scale Non-Bilinear Saddle Point Problems
    Le Thi Khanh Hien
    Renbo Zhao
    William B. Haskell
    Journal of Optimization Theory and Applications, 2024, 200 (1) : 34 - 67
  • [39] PRIMAL-DUAL FIRST-ORDER METHODS FOR AFFINELY CONSTRAINED MULTI-BLOCK SADDLE POINT PROBLEMS
    Zhang, Junyu
    Wang, Mengdi
    Hong, Mingyi
    Zhang, Shuzhong
    SIAM JOURNAL ON OPTIMIZATION, 2023, 33 (02) : 1035 - 1060
  • [40] An Inexact Primal-Dual Smoothing Framework for Large-Scale Non-Bilinear Saddle Point Problems
    Hien, Le Thi Khanh
    Zhao, Renbo
    Haskell, William B.
    JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 2024, 200 (01) : 34 - 67